Because of their complexity, methods for quantitation of anaphylaxis in large hypersensitive animals ( 1 ) are not readily adapted for the mouse. Consequently, the occurrence of anaphylaxis in mice is customarily determined using death as end point. Attempts have been made to estimate sublethal anaphylaxis in mice by observing the appearance of certain gross signs (ie., cyanosis and convulsions). Because of their subjectivity, these observations are inadequate. Kind ( 2) found that the rectal temperatures of mice surviving anaphylactic shock were markedly decreased.In our laboratory we consistently observed that hemoconcentration occurs in mice undergoing anaphylaxis. We observed signs of anaphylaxis similar to those described by Weiser et al. (3) . Immediately after challenge most of the mice became quiet and huddled in one position. The fur became ruffled and there was mild lacrimation. There soon followed partial paralysis of the limbs and marked cyanosis of the limbs, tail, and snout. Many animals experienced brief but violent convulsions separated by short periods of complete prostration. Nearly all mice which underwent convulsions died within 30 minutes. Nelson et al.(4) reported that the hematocrit of pooled mouse blood was markedly increased following fatal anaphylactic shock. The present study was designed to determine whether or not hematocrit change might be a suitable criterion for determination of anaphylaxis in mice.Methods. Sensitization and challenge. Fol-lowing randomization, female Swiss-Webster mice (22-27 g) were sensitized with 2 IP injections of 0.25 ml of antigen administered 4 days apart. Twelve days after receipt of the initial sensitizing injection the animals were challenged with an injection in the caudal vein of 0.25 ml of antigen. Two antigen preparations were employed: 1 ) alum precipitated bovine serum albumin* with Hemophilus pertussis vaccinet (APBA + HPV), each 0.25 ml containing 1 mg of bovine serum albumin suspended in 2.576 alum (pH 6.5 to 7.5) with lo9 to 1O1O H . pertussis vaccine cells; and, 2 ) the same as above except that H . pertussis vaccine was not included. This antigen preparation is referred to as APBA. A single challenge antigen preparation, bovine serum albumin with H . pertussis vaccine (BSA + HPV) was used. Each 0.25 ml of this preparation contained 4 mg of bovine albumin and 109 to 1010 cells H . pertussis vaccine. Controls included groups of mice sensitized with one or the other of the 2 sensitizing antigen preparations and challenged with saline, mice injected with saline instead of sensitizing antigen preparation and challenged with BSA + HPV, and groups given saline instead of both sensitization and challenge. Hematocrit determination: A modification of the microhematocrit method described by McGovern et al.( 5) was employed. Four-tenths mm I D microprecipitin capillary *Bovine serum albu,min, 3076, sterile, for use as t Pertussis vaccine (whooping cough bacterin), diluent in Rh testing. h o u r Labs. Wyeth Labs.